Harshil Patel

This article discusses the importance of ML Model Registries as a centralized repository for effective model management and documentation, highlighting its role in bridging the gap between experimentation and production activities.

Explores the fundamental differences between data-centric and model-centric approaches in machine learning, advocating for a hybrid approach that prioritizes data quality and infrastructure for robust AI systems.

Discusses the importance of machine learning reproducibility and highlights why Data Version Control (DVC) is superior to Git and Git-LFS for managing data in ML projects, covering key components like data, code, environment, and compute.

A guide on using MLflow for experiment tracking and managing the machine learning lifecycle, detailing its components like tracking, projects, models, and model registry, and emphasizing its benefits for reproducibility and collaboration.